ai-do
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional. It uses the
Writetool to save markdown files containing prompts to the local filesystem (ai-do-prompt.md). These files are meant for the user to copy-paste into subsequent sessions and do not contain executable scripts. - [SAFE]: While the skill mentions installing other skills via
npx skills add, it does not execute these commands itself. It provides the commands to the user as instructions within the saved documentation files. - [SAFE]: The skill uses
lscommands to check for locally installed skills in standard directories (skills/,~/.claude/skills/), which is a benign environmental check to improve its recommendations. - [SAFE]: References to external content (GitHub URLs) are limited to well-known repositories (e.g.,
github.com/lebsral/*) and are used only as references for the agent to fetch documentation about other skills.
Audit Metadata